DNA microarrays are capable of simultaneously evaluating the relative expression levels of thousands of genes, and have developed rapidly since their initial introduction. As a result, DNA microarrays are now one of the most preferred technologies for identifying early biomarkers of toxicity and disease. The outcome of microarray studies can be affected by many technical and instrumental factors, resulting in major criticism regarding lack of reproducibility and accuracy of the derived data. Although fluorescent dyes, surface chemistries, spotting robots, hybridization chambers, detection instruments, and data analysis tools have all undergone substantial development and refinement, the microarray substrate itself remains as a simple glass surface. In this proposal, we describe how replacement of the glass surface with a special-purpose optical transducer can provide quality control information on the interspot and intraspot density of microarray spots that is currently completely lacking from microarray analysis, while simultaneously amplifying the intensity of fluorescent labels used to quantify hybridized DNA. By providing information on spot variability, (representing a major source of error in microarray analysis), while at the same time increasing the signal-to-noise ratio for detection of weakly expressed genes (where microarray platforms currently face a disadvantage compared to other quantitative gene expression platforms), the proposed project represents a fundamental advance in microarray technology. The optical transducer used to provide these features is a 2-dimensional photonic crystal (PC) surface that is designed to provide optical resonances that enable high resolution label-free imaging detection of deposited microarray spots and up to 550x enhanced detection sensitivity of commonly used microarray fluorescent dyes. The PC is fabricated by a large-area nanoreplica molding process on plastic substrates that are attached to standard glass microscope slides for compatibility with existing spotting robots, hybridization chambers, and detection instruments. Recently, large area PC surfaces have been developed by the Cunningham Group at Illinois as multifunctional optical transducers that can be designed to produce narrow-wavelength electromagnetic resonances at any desired wavelength, featuring high intensity fields that extend evanescently into the media on the PC surface. The interaction of the optical resonance with adsorbed biomolecules results in a highly localized shift of the resonant wavelength that is used to quantify the density of adsorbed material without the use of fluorescent labels, enabling label-free images of deposited DNA microarray spots to be measured with 4 5m spatial resolution over a PC comprising the entire surface of a conventional microarray slide. A PC surface may also be designed so that the optical resonance coincides with the wavelength of a laser used to excite a fluorescent dye, thereby increasing the fluorescent output intensity relative to the intensity that would occur on an ordinary glass microarray slide, using an effect called Enhanced Fluorescence (EF). The EF effect has been shown to result in ~50x increase in the detected fluorescence signal using commercially available microarray laser scanning instruments, but can be further enhanced when the PC is designed to also incorporate an optical resonance at the emission wavelength of the fluorophore, resulting in an additional 10x gain in sensitivity. In the proposed effort, we plan for the first time to apply 2-dimensional PC surfaces that incorporate optical resonances for both label-free detection and EF to spotted gene expression microarrays. The label-free resonance will be utilized to quantify the density variability of deposited DNA spots, thereby providing a quality- control tool that is not currently available to researchers using spotted arrays. The label-free images of DNA spots will be used to quantify interspot and intraspot density variability, providing information that will be used to eliminate defective spots from further analysis or as a means for normalizing the detected signal from subsequent fluorescent measurements. The EF resonance will be applied to enhance the output of Cy5- labeled hybridized DNA, enabling gene expression analysis to be conducted with lower sample concentrations and the ability to observe gene expression at lower levels than has previously been possible. The project will enable collaboration between faculty in Electrical Engineering, who developed the PC and EF technology under NSF funding, and faculty in Crop Science, who manage the NSF Soybean Functional Genomics Center, thus allowing the technology to be fully tested and developed for large arrays. The benefits of the method will be statistically quantified on a 7680-element gene array with sufficient inter-chip and intra- chip replicates and controls to quantify sensitivity and quality control gains obtained from each independent PC transducer function. The resulting capability will be broadly applicable across a wide range of scientific research that utilizes microarrays for human, animal, and plant gene expression analysis. Analysis of soybean gene arrays was selected as an ideal testbed for the new sensor technology, as it will not require the safety and approval protocols for working with human DNA and human-derived test samples. PUBLIC HEALTH RELEVANCE: The proposed project seeks to develop a technology platform for providing high- sensitivity label-free detection of biomolecules and substantial amplification of fluorescence output on large-area, plastic based nanostructured surfaces called "photonic crystals." The goal is to incorporate photonic crystal surfaces into DNA microarray slides to provide label-free quality control of array spots and the ability to more easily detect and identify genes with low expression levels. The project is relevant for the development of gene-based diagnostic tests that are accurate, reliable, and able to identify genes at low concentration.